PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Sparse Multiview Methods for Classification of Musical Genre from Magnetoencephalography Recordings
Tom Diethe, Gabi Teodoru, Nick Furl and John Shawe-Taylor
Proceedings of the 7th Triennial Conference of European Society for the Cognitive Sciences of Music (ESCOM 2009) 2009.

Abstract

Classification of musical genre from audio is a well-researched area of music research. However to the authors’ knowledge no studies have been performed that attempt to identify the genre of music a person is listening to from recordings of their brain activity. It is believed that with the appropriate choice of experimental stimuli and analysis procedures, this discrimination is possible. The main goal of this experiment is to see whether it is possible to detect the genre of music that a listener is attending to from brain signals. The present experiment is focuses on Magnetoencephalography (MEG), which measures magnetic fields produced by electrical activity in the brain. We show that classification of musical genre from brain signals alone is feasible, but unreliable. We show that though the use of sparse multiview methods, such as Sparse Multiview Fisher Discriminant Analysis (SMFDA), we are able to reliably discriminate between different genres.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Brain Computer Interfaces
Theory & Algorithms
ID Code:5826
Deposited By:Tom Diethe
Deposited On:08 March 2010